Selected work

Representative Papers

Discovering Spatial Correlations between Earth Observations in Global Atmospheric State Estimation by using Adaptive Graph Structure Learning

Aug 11, 2025

Existing global atmospheric state estimation methods inadequately model spatial correlations between Earth observations and atmospheric fields. Method: We propose an adaptive spatiotemporal graph neural network (STGNN) framework that introduces an edge-sampling mechanism jointly guided by node-degree adaptivity and spatial-distance constraints to mitigate information loss and over-smoothing in graph learning; further, it jointly models meteorological observations and numerical weather prediction (NWP) gridded data to explicitly capture dynamic spatiotemporal dependencies. Contribution/Results: Evaluated on real-world observational data across East Asia, our model significantly outperforms state-of-the-art STGNN approaches—particularly in regions with sharp atmospheric transitions (e.g., frontal zones and typhoons), where forecast accuracy improves markedly. The framework establishes a new, interpretable, and robust paradigm for high-resolution atmospheric state estimation.

0 citationsRead paper

Mitigating Degree Bias in Graph Representation Learning with Learnable Structural Augmentation and Structural Self-Attention

Apr 21, 2025

Graph neural networks suffer from degree bias due to long-tailed degree distributions—high-degree nodes dominate message passing, while low-degree nodes suffer from insufficient information propagation and poor representation learning. To address this, we propose Degree-Fair Graph Transformer (DegFairGT), the first model integrating *intra-community role-aware learnable structural enhancement* with *structural self-attention*. It selectively incorporates role-similar non-neighboring nodes to enrich low-degree node representations while suppressing overloaded propagation from high-degree nodes. Additionally, we introduce a *p-step transition probability-based self-supervised regularization*, jointly optimizing structural fidelity and degree fairness. Extensive experiments across six benchmark datasets demonstrate that DegFairGT significantly improves degree fairness and consistently outperforms state-of-the-art methods on both node classification and clustering tasks.

0 citationsRead paper

Halal or Not: Knowledge Graph Completion for Predicting Cultural Appropriateness of Daily Products

Jan 10, 2025

Existing halal cosmetic classification methods rely solely on individual ingredient features, neglecting semantic relationships among ingredients, thereby limiting prediction accuracy. Method: This paper proposes a knowledge graph–based halal compliance prediction framework. It introduces knowledge graph completion to this domain for the first time, constructing a cosmetics–ingredients knowledge graph to explicitly model high-order semantic relations. We design a pretraining–fine-tuning residual graph attention network (RGAT) that jointly encodes multi-source ingredient interactions and cultural constraints. Contribution/Results: Evaluated on a dedicated halal cosmetic dataset, our method achieves a 6.2% absolute accuracy improvement over the state of the art. Results demonstrate that structured relational modeling significantly enhances reasoning for culturally sensitive products. This work establishes a novel paradigm for cross-cultural AI applications, bridging domain-specific knowledge representation with deep learning for regulatory compliance assessment.

0 citationsRead paper

Recent publications

Latest Papers

Discovering Spatial Correlations between Earth Observations in Global Atmospheric State Estimation by using Adaptive Graph Structure Learning

Aug 11, 2025

Existing global atmospheric state estimation methods inadequately model spatial correlations between Earth observations and atmospheric fields. Method: We propose an adaptive spatiotemporal graph neural network (STGNN) framework that introduces an edge-sampling mechanism jointly guided by node-degree adaptivity and spatial-distance constraints to mitigate information loss and over-smoothing in graph learning; further, it jointly models meteorological observations and numerical weather prediction (NWP) gridded data to explicitly capture dynamic spatiotemporal dependencies. Contribution/Results: Evaluated on real-world observational data across East Asia, our model significantly outperforms state-of-the-art STGNN approaches—particularly in regions with sharp atmospheric transitions (e.g., frontal zones and typhoons), where forecast accuracy improves markedly. The framework establishes a new, interpretable, and robust paradigm for high-resolution atmospheric state estimation.

0 citationsRead paper

Mitigating Degree Bias in Graph Representation Learning with Learnable Structural Augmentation and Structural Self-Attention

Apr 21, 2025

Graph neural networks suffer from degree bias due to long-tailed degree distributions—high-degree nodes dominate message passing, while low-degree nodes suffer from insufficient information propagation and poor representation learning. To address this, we propose Degree-Fair Graph Transformer (DegFairGT), the first model integrating *intra-community role-aware learnable structural enhancement* with *structural self-attention*. It selectively incorporates role-similar non-neighboring nodes to enrich low-degree node representations while suppressing overloaded propagation from high-degree nodes. Additionally, we introduce a *p-step transition probability-based self-supervised regularization*, jointly optimizing structural fidelity and degree fairness. Extensive experiments across six benchmark datasets demonstrate that DegFairGT significantly improves degree fairness and consistently outperforms state-of-the-art methods on both node classification and clustering tasks.

0 citationsRead paper

Halal or Not: Knowledge Graph Completion for Predicting Cultural Appropriateness of Daily Products

Jan 10, 2025

Existing halal cosmetic classification methods rely solely on individual ingredient features, neglecting semantic relationships among ingredients, thereby limiting prediction accuracy. Method: This paper proposes a knowledge graph–based halal compliance prediction framework. It introduces knowledge graph completion to this domain for the first time, constructing a cosmetics–ingredients knowledge graph to explicitly model high-order semantic relations. We design a pretraining–fine-tuning residual graph attention network (RGAT) that jointly encodes multi-source ingredient interactions and cultural constraints. Contribution/Results: Evaluated on a dedicated halal cosmetic dataset, our method achieves a 6.2% absolute accuracy improvement over the state of the art. Results demonstrate that structured relational modeling significantly enhances reasoning for culturally sensitive products. This work establishes a novel paradigm for cross-cultural AI applications, bridging domain-specific knowledge representation with deep learning for regulatory compliance assessment.

0 citationsRead paper